基于互補性的運動模糊圖像復原方法
首發時間:2017-05-02
摘要:圖像復原的目的是將原始圖像從觀測到的降析圖像中恢復出來,在多幀圖像復原算法中,輸入圖像幀的選擇是很重要的問題,輸入圖像選擇得好壞,直接影響最終的復原效果。本文針對多幀圖像復原算法中的選幀問題,提出了一種基于互補性的運動模糊圖像復原算法,首先根據圖像紋理特性進行互補性計算,采用基于迭代的最大互補信息序列的選幀方法,挑選出最具互補性的多幀模糊圖像,之后依據基于隱變量的稀疏建模的圖像盲復原算法進行模糊核與潛在圖像估計。實驗表明,該方法的復原效果較好,優于單一的多幀圖像復原方法。
For information in English, please click here
Multi deblurring method based on complementarity
Abstract:The purpose of image restoration is to recover the original image from the observed image. In the multi-frame image restoration algorithm, the selection of the input image frames is a very important problem. The input image's quality is good or bad, directly affect the final deblurring result. In this paper, a motion-blurred image restoration method based on the image's complementarity is proposed for the framing problem in multi-frame image restoration algorithm. First, the complementarity calculation is carried out according to the image texture characteristics. The most complementary multi-frame blurred image is selected by using the frame selection method based on the iterative maximum complementary information sequence. Then, using a coupled adaptive sparse prior model to estimate the blur kernel and the potential image. Experiments show that the result by using the proposed method is better than the traditional multi-image deblurring method.
Keywords: image restoration multi deblurring complementary
基金:
論文圖表:
引用
No.4726393119355514****
同行評議
共計0人參與
勘誤表
基于互補性的運動模糊圖像復原方法
評論
全部評論